ARTICLE IN BRIEF
✓ The first complete map of structural variation in the human genome, reported in 2006, show many more copy number variations — covering two to three times more nucleotides than those governed by single base-pair changes — as a potential explanation for seemingly random diseases, notably those of the CNS and brain.
If the Human Genome Project was biomedicine's “moon-shot,” new findings suggest that our genetic topography more closely resembles that of Mars — with far more complexity and diversity than the sparse lunar landscape.
Writing in the Mar. 15 New England Journal of Medicine, noted geneticist James R. Lupski, MD, PhD, points to the first complete map of structural variation in the human genome, reported in 2006, showing many more copy number variations (CNV) — covering two to three times more nucleotides than those governed by single base-pair changes – as a potential explanation for seemingly random diseases, notably those of the CNS and brain.
Finding so many CNVs in the genome could lead to a fundamental reappraisal of what causes many incurable diseases, said Dr. Lupski, professor and vice chair of molecular and human genetics at Baylor College of Medicine in Houston, TX.
DNA microchip arrays have shown that large portions of the genome are often duplicated or deleted abnormally, with gains and losses of genetic information. These structural rearrangements and subsequent changes in copy number of genes alter the regulation of proteins and upset the normal balance in the body. If defects are great enough, scientists believe and some studies have shown they can result in diseases that previously defied genetic classification. These diseases may be influenced by the number of copies of certain key genes, and the mutations that become active throughout a person's life.
“The demonstration of aberrations in gene dosage (the number of copies of a given gene in a cell or nucleus) as a disease mechanism opens the way to new, more easily developed approaches to treatment in which the goal is not to correct abnormal or mutant proteins but instead to modify their abnormal dosage,” Dr. Lupski wrote.
These CNV findings will ultimately change investigations into the causes of human genetic diseases, he said.
“From now on, all genetic linkage and association studies should incorporate an evaluation of the variations in copy number in the study population to determine whether an individual variation in copy number, rather than a single-nucleotide polymorphism, might be responsible for the trait being investigated,” he said.
“Neurons don't turn over as readily as other cells,” he noted. “Subtle defects in neurons may cause subtle abnormalities that may gradually have a great impact. We have to expand our efforts beyond base-pair changes. There can be a gentle decline, and all of the traits don't have to be inherited, they can be de novo mutations.”
In 2006, Dr. Lupski and graduate student Jennifer Lee described neurodevelopmental, neurodegenerative, behavioral, and psychiatric disorders known or believed to result from large genomic changes (Neuron 2006;52:103–121). (See “Diseases Linked to Copy Number Variation.”)
CNV could help explain why people can inherit a gene but do not show signs of disease until later in life, Dr. Lupski told Neurology Today in a telephone interview. It also helps explain sporadic or spontaneously occurring cases of diseases with a genetic component, he said.
“It appears that rather than resulting from simple DNA changes, many mendelian and complex traits, as well as sporadic diseases, may indeed result from structural variation.”
SINGLE VS. MULTIPLE MUTATIONS
To date, the molecular medicine model that is promulgated in every medical school is based on sickle cell disease. In that disease, the predominant type of mutation is a base-pair change, which alters the coding sequence and results in the synthesis of a mutant protein. In this model, genetic variation between individuals and between populations arises from variant bases, also known as single nucleotide polymorphisms (SNPs).
“Many variations in copy number are probably benign, but specific variations are associated with common mendelian (single-locus) conditions such as color blindness and Charcot-Marie-Tooth disease,” Dr. Lupski wrote. “Variation in copy number can also influence susceptibility to complex diseases such as lupus glomerulonephritis and to infection with the human immunodeficiency virus, Parkinson's disease, Alzheimer's disease, and Crohn's disease.”
For instance, French researchers identified duplications of the amyloid precursor protein gene locus as the cause of early-onset Alzheimer disease in five families (Nat Genet 2006;38:24–26).
Initially reported in 2003 and confirmed in 2006, scientists have found that people with CNVs that promote the activity of the alpha-synuclein gene have 1.5 times greater risk of developing Parkinson disease (Science 2003;302:841); (JAMA 2006;296:661–670).
Dr. Lupski pointed to a study published Mar. 28 in the online journal Public Library of Science ONE, led by Arthur Beaudet, MD, chair of the Baylor College of Medicine department of molecular and human genetics.
In 2,513 postnatal samples from children with a variety of clinical phenotypes, the team evaluated versions of array Comparative Genomic Hybridization (a-CGH), a powerful molecular tool to detect genomic imbalances using Chromosomal Microarray Analysis (CMA). They found the technique, especially the latest versions, is extremely sensitive in detecting potential genetic abnormalities. They were evaluating the technique for accuracy in detecting mutations and copy number variations, not specific diseases.
While it did not identify a problem in all children, there was a 5 percent to 12 percent chance that it could identify an abnormality with different disabilities where previous chromosomal testing did not. A study published online Mar. 15 in advance of the print edition of Science also illuminates the potential of the new technology in identifying genomic variations in families with autism.
Researchers at Cold Spring Harbor Laboratory in New York reported the largest percentage of CNVs in families with one autistic child. Single autistic children are called sporadic or spontaneously occurring cases, unlike families with more than one autistic child, a pattern indicating single-point genetic inheritance.
Ten percent of children with sporadic autism had de novo CNVs versus 1 percent of controls with no autism and only 2 percent of families with multiple autistic children.
The price of DNA microarrays is falling rapidly, said Dr. Lupski. But unfortunately, when most clinicians see that there's no family history of a disease in a patient with symptoms resembling those with a known genetic cause, they often start looking for other causes, he noted.
“Clinicians want to see a family history — it's really that simple.” This is especially true for neurological disorders where de novo mutations and sporadic changes occur with greater frequency, he added.
“What we're discovering about these more subtle mutations is terrific for clinicians, especially for pediatric neurologists. You don't have to be a geneticist to use pattern expression testing. Now you can do genome analysis for CNVs relatively inexpensively. Some of the tests are available for $1,000 to $1,500. If you think it's possibly chromosomal, you can test for hundreds of different disorders in that price range,” he said.
Thomas Bird, MD, professor of medicine, neurology, and medical genetics at the University of Washington, in Seattle, also said CNV findings have direct clinical applications.
“My view is that neurogenetics is becoming more important as people start to realize that there are all these other, more subtle genetic alterations — micro-deletions and duplications — behind some neurological disorders, and not just single-point mutations. This isn't necessarily new, but the number is becoming larger and larger,” he told Neurology Today in a telephone interview.
“At some point down the line, this line of research will result in treatment advances. The technology is getting cheaper to use. Although the cost of scans varies from gene to gene, in fact the cost is all over the genetic map, everyone hopes it will become much less expensive. I suspect that it will reach around $500 per gene before too long.”
DISEASES LINKED TO COPY NUMBER VARIATION
The following neurological disorders have already been linked to structural rearrangements and copy number variation:
* Alzheimer disease
* Parkinson disease
* Spinal muscular atrophy (SMA)
* Neurofibromatosis type 1
* Charcot-Marie-Tooth disease type 1A
* Williams-Beuren syndrome
* Miller-Dieker syndrome and lissencephaly
* Angelman and Prader-Willi syndromes
* Smith-Magenis syndrome